All publications mentioned throughout this application are fully incorporated herein by reference, including all references cited therein.
Lipids in general are the building blocks of life. They are used as building blocks of membranes, cells and tissues, as energy sources, either immediate or stored, as precursors to a variety of other bio-molecules, as well as biochemical signals. In all biochemical processes lipids have an important role.
Many lipids, and especially triglycerides, are consumed in the human nutrition on a daily basis. In most cases, these lipids are metabolized and used for energy storage, precursors for biosynthesis of other lipids or bio-molecules. Whatever the fate of the lipids in the metabolic pathways, during and after their consumption, they interact with other nutrients or their metabolic products.
In human milk, and in most infant formulas, about 50% of the dietary calories are supplied to newborns as fat. More than 98% of this milk fat is in the form of triglycerides, which contain saturated and unsaturated fatty acids esterified to glycerol.
Fatty acids in human milk fat have a highly specific positional distribution on the glycerol backbone. This specific configuration is known to have a major contribution to the efficiency of nutrient absorption.
Palmitic acid (C16:0) is the predominant saturated fatty acid (SAFA), constituting 20-25% of the fatty acids in mature human milk. 70-75% of this fatty acid are esterified at the sn-2 position of the triglycerides. In contrast, palmitic acid present in vegetable oils, which are most commonly used in the manufacture of infant formulas, is esterified at the sn-1 and sn-3 positions, while the sn-2 position is predominantly occupied by unsaturated fatty acids.
Linoleic (18:2) and linolenic (C18:3) acids cannot be synthesized in animal tissues and must be obtained from the diet, i.e. ultimately from plants. There is an absolute requirement for these so called “essential fatty acids” for growth, reproduction and good health. In triglycerides in human mother milk, 5-20% of the total C18:3 and 20-23% of the total C18:2 are esterified at the sn-2 position of the glycerol backbone [Lopez-Lopez A. (2002) European Journal of Clinical Nutrition; 56:1242-54, Innis S. M. (1994) Lipids; 29:541-5].
Triglyceride Digestion by the Infant
The triglyceride digestive process of the neonate is complex. It is initiated by a gastric phase catalyzed by gastric or lingual lipase [Hamosh M. (1990) Nutrition; 6:421-8]. This initial lipolysis allows maximal activity of pancreatic colipase-dependent lipase during the intestinal phase of digestion. The pancreatic lipase system attacks the triglyceride with a high degree of positional specificity. Lipolysis occurs predominantly at the sn-1 and sn-3 positions, yielding two free fatty acids and a 2-monoglyceride [Mattson F. H. & Beck L. H. (1956) J. Biol. Chem.; 219:735-740]. Monoglycerides are well absorbed independent of their constituent fatty acid. In contrast, the absorption of free fatty acids varies greatly, depending on their chemical structure. Mono and polyunsaturated fatty acids are well absorbed, as are saturated fatty acids of 12 carbons or less in chain length. The coefficient of absorption of free long chain saturated fatty acids i.e. palmitic acid is relatively low [Jensen C. et al. (1988) Am. J. Clin. Nutr.; 43:745-51], due in part to a melting point above body temperature (˜63° C.) and the tendency of these fatty acids to form hydrated fatty acid soaps with minerals such as calcium or magnesium at the pH of the intestine [Small D. M. (1991) Annu. Rev. Nutr.; 11:413-434].
Several studies have demonstrated the preferential absorption of palmitic acid when present at the triglyceride sn-2 position [Lien E. L. et al. (1997) J. Ped. Gastr. Nutr.; 52(2):167-174; Carnielli V. P. et al. (1995) Am. J. Clin. Nutr.; 61:1037-1042; Innis S. M. et al. (1993) Am. J. Clin. Nutr.; 57:382-390; Filer L. J. et al. (1969) J. Nutr.; 99:293-8]. Studies comparing the palmitic acid absorption of human milk and formulas conclude that the absorption of palmitic acid is higher in human milk [Chappel J. E. et al. (1986) J. Pediatr.; 108:439-447; Hanna F. M. et al. (1970) Pediatr.; 45:216-224; Tommarelli R M., et al. (1968) J. Nutr.; 95:583-90]. The greater absorption of fat and calcium in breast-fed infants compared with those fed formula has been ascribed to two factors: the presence of a lipolytic enzyme (the bile salt-stimulated lipase) in breast milk and the relatively high proportion of palmitic acid at the sn-2 position of the triglyceride [Hernell O. et al. (1988) Perinatal Nutrition. New York: Academic Press.; 259-272; Wang C. S. et al. (1983) J. Biol. Chem.; 258:9197-9202]. Higher palmitic acid absorption was obtained with formulas rich in palmitic acid esterified in the sn-2 position of the triglycerides, than with those containing palmitic acid predominantly esterified in the sn-1,3 positions [López-López A. et al. (2001) Early Hum. Deo.; 65:S83-S94].
A study comparing the absorption of fat and calcium by infants fed a formula containing a blend of palm olein and soy oil (high levels of palmitic acid at the sn-1,3 positions) and a formula containing a blend of soy oil and coconut oil (low levels of palmitic acid) showed that the mixture of palm olein and soy oil, although providing the proportion of palmitic and oleic acids similar to those of human milk fat, was less absorbed [Nelson S E. et al. (1996) Am. J. Clin. Nutr.; 64:291-296]. Another study showed that fat absorption in infants fed formula containing lard was reduced when the high proportion of sn-2 palmitin in lard was reduced to 33% by chemical randomization [Filer (1969) id ibid.].
The composition of monoglycerides absorbed from the intestinal lumen is important to the fatty acid distribution of circulating lipids because about 70% of the fatty acids absorbed as sn-2 monoglycerides are conserved in the original position during re-esterification to form triglycerides in the intestinal cells [Small (1991) id ibid.].
Studies in piglets provided evidence that palmitic acid, when absorbed from milk or formula with rearranged triglycerides as a sn-2 monoglyceride, is conserved through the process of triglyceride reassembly in the enterocyte and secretion in plasma lipoprotein triglycerides [Innis S. M. et al. (1995) J. Nutr.; 125:73-81]. It has also been shown that the distribution of saturated fatty acids in human milk and infant formula is a determinant of the fatty acid distribution of infant plasma triglycerides and phospholipids [Innis S. M. et al. (1994) Lipids.; 29:541-545].
During the first year of life an infant's birth weight triples and the length is increased by 50%. To meet the requirements of their rapidly expanding skeletal mass, growing infants require a bioavailable source of calcium. For formula-fed infants, availability of calcium depends on the composition of the formula [Ostrom K. M. et al. (2002) J. Am. Coll. Nutr.; 21(6):564-569].
As mentioned above, the digestion of triglycerides involves lipolysis at the sn-1 and 3 positions and formation of free fatty acids and 2-monoglycerides. When palmitic acid is located at the sn-1,3 positions, as is the case in most infant formulas, it is released as free fatty acid which tends to form insoluble calcium soaps. In contrast, palmitic acid esterified to the sn-2 position, as in human milk, is unavailable to form calcium soaps [Small (1991) id ibid.].
Several studies have shown a correlation between formulas containing high levels of palmitic acid situated at the sn-1,3 positions of the triglyceride and reduction in calcium absorption [Nelson S. E. et al. (1998) J. Amer. Coll. Nutr.; 17:327-332; Lucas A. et al. (1997) Arch. Dis. Child.; 77:F178-F187; Carnielli V. P. et al. (1996) J. Pediatr. Gastroenterol. Nutr. 23:553-560; Ostrom (2002) id ibid.;
Hanna (1970) id ibid.]. In addition, it was shown that dietary triglycerides containing palmitic acid predominantly at the sn-2 position, as in human milk, have significant beneficial effects on the intestinal absorption of fat and calcium in healthy term infants as well as in preterm infants [Carnielli (1996) id ibid.; Carnielli (1995) id ibid.; Lucas (1997) id ibid.]. Infants fed a formula containing high levels of palmitic acid at the sn-1,3 positions showed greater fecal excursion of calcium and, hence, a lower percentage absorption of calcium compared to infants fed a formula containing low levels of palmitic acid [Nelson (1996) id ibid.]. Fecal excretion of calcium was closely related to the fecal excretion of fat. This study also showed that urinary phosphorus excretion increased and phosphorus retention decreased when infants were fed the formula containing high levels of palmitic acid at the sn-1,3 positions. These findings presumably reflect lower availability of calcium for deposition in bones.
The impact of soap formation on calcium absorption can be significant. Many infant formulas contain sufficient saturated fatty acids to form soaps with virtually all the calcium available.
Another important issue which is associated with formula feeding is constipation in both term and preterm infants which, in the latter, can lead to life threatening complications. By contrast, constipation is rare in breast fed term infants. A study comparing breast fed and formula fed infant stool hardness and composition showed that calcium fatty acid soaps are positively correlated to stool hardness. Stools from formula-fed infants were significantly harder than those of the breast-fed infants suggesting different handling of saturated fatty acids [Quinlan P T. et al. (1995) J. Pediatr. Gastr. and Nutr.; 20:81-90].
In an attempt to overcome the decreased calcium absorption and hard stool phenomena, infant formula manufacturers tend to deviate from the fatty acid profile by replacing palmitic acid with lauric acid and, in some cases, by increasing the polyunsaturated fatty acid content. Studies have shown that fatty acid composition of the diet influences the fatty acid composition of developing infant tissue [Widdowson E. M. (1975) Br. Med. J.; 1:633-5; Carlson S. E. et al. (1986) Am. J. Clin. Nutr.; 44:798-804; Innis S. M. et al. (1990) Am. J. Clin. Nutr.; 5:994-1000; Koletzko B. et al. (1989) Eur. J. Pediatr.; 148:669-75] and thus the lipoprotein and lipid metabolism differ between breast-fed and formula-fed infants [Putnam J. C. et al. (1982) Am. J. Clin. Nutr.; 36:106-114; Innis S. M. et al. (1992) Am. Coll. Nutr.; 11:63S-8S; Van Biervliet J. P. et al. (1981) Acta. Paediatr. Scand.; 70:851-6].
Innis and colleagues [Innis (1993) id ibid.], when comparing three formulas containing similar amounts of saturated fatty acids—C8-C14, C16 from palm oil (predominantly in the sn-1,3 positions), or C16 from synthesized triglyceride (predominantly in the sn-2 position)—showed that the chain length of saturated fatty acids in infant formula influences the metabolism of the dietary oleic, linoleic and alpha-linolenic acids. This study also showed that the sn-2 configuration of C16 in human milk triglycerides seems to have unique properties that extend beyond absorption. These include effects on HDL and cholesterol concentrations, and the cholesterol ester fatty acid composition.
EP 0 209 327 describes a substitute milk fat composition which is suitable for use as replacement fat in infant formulations. In the claimed fat composition at least 43.5% of the total saturated fatty acids residues are bound to the sn-2 position of the triglycerides. Additionally, at least 50% of the fatty acids occupying the sn-2 position are saturated. The fat composition of this patent is produced from an oil or fat characterized by high levels of saturated fatty acids, specifically palmitic acid.
EP 0 495 456 also discloses substitute human milk fat compositions. The claimed compositions have at least 40% of the total saturated fatty acids located at the sn-2 position of the triglycerides contained in the composition. These fat composition are also characterized by comprising 0.2-7% of linolenic acid moieties (C18:3, ω-3), 70% of which are bonded at the sn-1 and 3-positions of the glycerol moieties. The fat composition of EP 0 495 456 is also produced by transesterifications of a triglyceride source rich in saturated fatty acids, specifically palmitic acid. This source, such as top fraction of palm oil, can contain palmitic acid at levels higher than 80% and even 90%.
U.S. Pat. No. 5,658,768 discloses a multiple-step process for preparing triglyceride compositions in which more than 40% of the saturated fatty acid moieties are at the sn-2 position. Many of the steps involve enzymatic modifications.
Applicant's WO2005/036987 describes a fat composition which can be used to create human milk fat substitutes which have great similarity to human milk fat and/or can be used at relatively low levels to yield such human milk fat substitutes. This composition is also produced starting from triglycerides sources rich in palmitic acid.
While the fat compositions described above can be used in the preparation of human milk fat substitutes with high level of similarity to human milk fat (HMF), which in turn can be used in the creation of advanced infant formulas, the resulting HMF substitutes may be extremely expensive, making them impractical for the use in the preparation of commercial infant formulas.
The inventors of the current invention have addressed the issue of cost by producing a unique fat composition, described in said WO2005/036987, offering HMF substitutes with still high levels of similarity to HMF, by using relatively low levels of the synthetic fat-base composition of that invention in the total fat blend used in the infant formula.
Another approach of overcoming the cost obstacle provides HMF substitutes that offer a relatively high degree of the total saturated fatty acids located at the sn-2 position, yielding fat compositions with more than 40% of their total saturated fatty acids positioned at the sn-2 position (but less than 50%). Such HMF substitutes were produced by high degree of dilution of fat-base compositions, such as described by EP 0 209 327 or EP 0 495 456, with simple and cost-effective vegetable oils. These blends, however, have a lower ratio of sn-2 palmitic to total palmitic (˜40-43%) than can be found in HMF (˜70%). Nevertheless, clinical studies [Lien et al. Journal of Pediatric Gastroenterology and Nutrition 1997, 167-74] have shown that these blends offer a higher calcium and fatty acid intake than infant formulas based on mere blends of vegetable oils. Substitutes which are very close to human milk fat are described in applicant's said WO2005/036987 [InFat 1, Table 1]. In all cases, these HMF substitutes, characterized in having at least 40% of their saturated fatty acids located at the sn-2 position, are obtained by diluting fat-bases with high levels of sn-2 palmitic acid, as well as a high ratio of sn-2 palmitic to total palmitic acid, with a variety of vegetable triglycerides to obtain a blend in which at least 40% of the saturated fatty acids are bound at the sn-2 position. This ratio is usually between 40 to 45%. The fat bases used in the preparation of these HMF substitutes are produced by interesterifying triglycerides rich in palmitic acid with an excess of a mixture of free fatty acids, rich in oleic acid.
It is the object of the present invention to provide novel fat-base compositions, which would possess the benefits of good HMF mimics, yet be cost effective to produce.
It is a further object of the invention to provide novel processes for the low-cost production of HMF mimetic fat bases and fat blend suitable for use in various infant formulas.
These and other objects of the invention will become apparent as the description proceeds.